An inclined shaft data processing method, processing system, electronic equipment and readable storage medium

Abstract

This invention discloses a method, system, electronic device, and readable storage medium for processing deviated well data. The method includes: acquiring deviated well drilling sequence data; processing the deviated well drilling sequence data to obtain initial model data; generating a deviated well BIM model based on the initial model data; and sending the deviated well BIM model to a client for display. This invention constructs a mapping model between deviated well drilling sequence data and the rock mechanical properties of the deviated well. It generates parameters reflecting the rock mechanical properties of the deviated well using the drilling sequence data, effectively improving the efficiency of obtaining rock mechanical parameters. Simultaneously, this invention constructs a multi-target recognition model that can memorize hidden state information at multiple times, effectively improving the accuracy of rock mechanical parameter generation. Furthermore, this invention constructs a deviated well BIM model, effectively presenting rich deviated well information, facilitating rapid assessment of construction risks by relevant personnel and timely implementation of risk prevention and control measures.

Description

Technical Field

[0001] This invention belongs to the field of data processing in water conservancy and hydropower engineering, and specifically relates to a method, system, electronic device and readable storage medium for processing data of inclined wells. Background Technology

[0002] Inclined shafts / vertical shafts are an important component of hydropower projects, undertaking multiple functions such as water diversion, pressure regulation, and gate opening and closing. Due to the limited working space during inclined shaft construction, the consequences of rockfalls or internal collapses during operations can be incalculable. Therefore, studying the rock mechanics properties of inclined shafts and obtaining their rock mechanics parameters is particularly important.

[0003] Currently, obtaining the rock mechanical properties of inclined shafts mainly relies on rock samples collected from exploration boreholes. These samples are then tested in a laboratory to obtain the mechanical parameters characteristic of the rock samples. By taking samples from specific points and analyzing them across the entire inclined shaft, the rock mechanical parameters can be obtained. While this method can obtain the rock mechanical parameters of the inclined shaft, it lacks timeliness due to the need for testing. If these rock mechanical parameters are used to guide operations in later stages of construction, it will lead to extended construction periods, high costs, and low efficiency. Conversely, if these parameters are not used to guide operations in later stages of construction, safety risks cannot be guaranteed.

[0004] Therefore, there is an urgent need to develop a method that can obtain the rock mechanics parameters of inclined shafts in water conservancy and hydropower projects in a timely manner to guide the subsequent construction of inclined shafts. Summary of the Invention

[0005] The present invention aims to improve the efficiency of inclined shaft operations and ensure the inherent safety of inclined shafts by proposing an inclined shaft data processing method, processing system, electronic equipment, and readable storage medium.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: a method for processing deviated well data, the method comprising: Acquire deviated well drilling sequence data; the deviated well drilling sequence data includes at least time, location, drilling pressure, torque, rotational speed, drilling rate, and displacement; The drilling sequence data of the deviated well is processed to obtain initial model data; the initial model data includes at least location, lithology, rock mass compressive strength, rock mass integrity and surrounding rock grade. Generate a BIM model of the inclined shaft based on the initial model data; The BIM model of the inclined shaft is sent to the client for display.

[0007] Furthermore, the processing of the deviated well drilling sequence data to obtain initial model data specifically includes: The deviated well drilling sequence data is input into a pre-trained first recognition model, and the initial model data is output based on the pre-trained first recognition model. The pre-trained first recognition model adopts a multi-target recognition model.

[0008] Furthermore, the multi-target recognition model includes an input layer, an encoding layer, a first multi-tower network layer, a memory layer, a second multi-tower network layer, a decoding layer, and an output layer; The input layer is used to input the deviated well drilling sequence data and transmit the deviated well drilling sequence data to the encoding layer; The encoding layer is used to encode the deviated well drilling sequence data into a first feature vector, and transmit the first feature vector to the first multi-tower network layer; The first multi-tower network layer includes three tower networks, which are used to process the first feature vector respectively to generate a first intermediate vector, a second intermediate vector and a third intermediate vector corresponding to lithology, rock mass compressive strength and rock mass integrity, and transmit the first intermediate vector, the second intermediate vector and the third intermediate vector to the memory layer; The memory layer is used to remember the first intermediate vector, the second intermediate vector and the third intermediate vector corresponding to multiple time points, to obtain a memory vector, and then transmits the memory vector to the second multi-tower network layer; The second multi-tower network layer includes three tower networks, which are used to process the memory vectors respectively to generate a first decoding vector, a second decoding vector, and a third decoding vector corresponding to lithology, rock mass compressive strength, and rock mass integrity, and transmit the first decoding vector, the second decoding vector, and the third decoding vector to the decoding layer; The decoding layer is used to decode the first decoding vector, the second decoding vector, and the third decoding vector to obtain the lithology, the rock mass compressive strength, the rock mass integrity, and the surrounding rock grade, and transmits the lithology, the rock mass compressive strength, the rock mass integrity, and the surrounding rock grade to the output layer; The output layer is used to output the lithology, the rock mass compressive strength, the rock mass integrity, and the surrounding rock grade.

[0009] Furthermore, the decoding layer includes four decoding sub-modules. Decoding the first decoding vector, the second decoding vector, and the third decoding vector to obtain the lithology, the rock mass compressive strength, the rock mass integrity, and the surrounding rock grade specifically includes: The first decoding vector is decoded using the first decoding submodule to obtain the lithology; The second decoding vector is decoded using the second decoding submodule to obtain the compressive strength of the rock mass; The third decoding vector is decoded using the third decoding submodule to obtain the integrity of the rock mass; The fourth decoding submodule is used to decode the first decoding vector, the second decoding vector, and the third decoding vector to obtain the surrounding rock grade.

[0010] Furthermore, the step of generating the inclined shaft BIM model based on the initial model data specifically includes: Acquire first model data and second model data, wherein the first model data is constructed based on the geometric parameters of the inclined shaft and the second model data is constructed based on the geological parameters of the inclined shaft; The initial model data is used as the first layer data of the BIM model, the first model data is used as the second layer data of the BIM model, and the second model data is used as the third layer data of the BIM model. The first layer data, the second layer data and the third layer data are superimposed to generate the inclined shaft BIM model.

[0011] Furthermore, the acquisition of the first model data specifically includes: Acquire 3D point cloud data collected by the advanced probe; The second recognition model, which is pre-trained, is used to identify three-dimensional point cloud data and obtain the geometric parameters of the inclined shaft. The geometric parameters of the inclined shaft include at least the location and the outline of the inclined shaft. The first model data is constructed based on the geometric parameters of the inclined shaft.

[0012] Furthermore, the acquisition of the second model data specifically includes: Acquire imaging data collected by the advanced probe; The pre-trained third recognition model is used to identify the imaging data and obtain the geological parameters of the inclined well. The geological parameters of the inclined well include the geological body type, occurrence and lithology at different locations. Second model data was constructed based on the geological parameters of the inclined shaft.

[0013] Furthermore, this invention also proposes a deviated well data processing system, which includes: The acquisition module is used to acquire deviated well drilling sequence data; The processing module is used to process the deviated well drilling sequence data to obtain initial model data; the initial model data includes at least location, lithology, rock mass compressive strength, rock mass integrity, and surrounding rock grade; The generation module is used to generate a BIM model of the inclined shaft based on the initial model data; The sending module is used to send the inclined shaft BIM model to the client for display.

[0014] Furthermore, the present invention also proposes an electronic device, including a memory and a processor, wherein the memory stores a computer program, characterized in that the processor executes the computer program to implement the steps of the method as described above.

[0015] Furthermore, the present invention also proposes a computer-readable storage medium for storing computer instructions, characterized in that the computer instructions, when executed by a processor, implement the steps of the method as described above.

[0016] Furthermore, the present invention also provides a computer program product that, when executed by a processor, implements the steps of the method as described above.

[0017] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention constructs a mapping model between deviated well drilling sequence data and deviated well rock mechanical properties. It generates parameters reflecting the rock mechanical properties of the deviated well through the deviated well drilling sequence data, which effectively improves the efficiency of obtaining rock mechanical parameters. At the same time, it guides the deviated well advance detection based on the generated rock mechanical parameters, which improves the accuracy of advance detection.

[0018] (2) The present invention constructs a multi-target recognition model based on neural networks. This multi-target recognition model can memorize the hidden state information at multiple times. Through this multi-target recognition model, the drilling sequence data can be identified, thereby generating rock mechanical parameters, which effectively improves the accuracy of generation.

[0019] (3) Based on the initial model data reflecting the rock mechanics parameters of the inclined shaft, the first model data reflecting the geometric parameters of the inclined shaft, and the second model data reflecting the geological parameters of the inclined shaft, the present invention constructs a BIM model of the inclined shaft, which can effectively present rich information about the inclined shaft, making it easier for relevant personnel to quickly assess construction risks and take timely risk prevention and control measures. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the deviated well data processing method provided in an embodiment of the present invention.

[0021] Figure 2 This is a schematic diagram of the first recognition model provided in an embodiment of the present invention.

[0022] Figure 3 This is a schematic diagram of the inclined well data processing system provided in an embodiment of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] See Figure 1 This invention proposes a method for processing deviated well data, the method comprising: S1: Obtain the deviated well drilling sequence data; the deviated well drilling sequence data includes at least time, location, drilling pressure, torque, rotational speed, drilling speed, and displacement.

[0025] Inclined shaft construction typically employs a reverse shaft drilling technique. This involves drilling directional holes from top to bottom in the horizontal section of the water diversion system according to the designed location, azimuth, and inclination angle, with a diameter typically ranging from 20 to 40 cm. After the directional holes are drilled, a reverse shaft drilling rig is used to pull the shaft upwards, creating a chute well with a diameter of 2 to 2.4 m. Sensors for drilling pressure, rotational speed, torque, drilling speed, and displacement are installed on the drilling rig used for drilling the directional holes to collect real-time data during the inclined shaft drilling sequence. After the chute well is completed, a forward probe is used to conduct an inclined shaft forward probe from top to bottom within the chute well.

[0026] This invention transmits the real-time acquired deviated well drilling sequence data to an integrated control platform, which then processes the deviated well drilling sequence data. The deviated well drilling sequence data includes at least time T, position POS, drilling pressure DP, torque TOR, rotational speed RSP, drilling speed DSP, and displacement DIS.

[0027] S2: Process the drilling sequence data of the deviated well to obtain initial model data; the initial model data includes at least the location POS, lithology LITH, rock mass compressive strength RC, rock mass integrity KV, and surrounding rock grade RANK.

[0028] To obtain rock mechanics parameters characterizing the rock mass mechanical properties of the deviated well based on the deviated well drilling sequence data, a mapping model can be pre-constructed between the deviated well drilling sequence data and the rock mass mechanical parameters of the deviated well, including lithology LITH, rock mass compressive strength RC, and rock mass integrity KV, as well as an evaluation model between lithology LITH, rock mass compressive strength RC, rock mass integrity KV and surrounding rock grade RANK. Based on the mapping model and the evaluation model, the deviated well drilling sequence data can be converted into rock mechanics parameters.

[0029] In one embodiment, the mapping model adopts a regression model, namely, a first regression model, a second regression model, and a third regression model trained based on historical data to obtain the relationships between lithology LITH, rock mass compressive strength RC, rock mass integrity KV, and drilling pressure DP, torque TOR, rotational speed RSP, drilling speed DSP, and displacement DIS, respectively. ; ; ; in, , , The first, second, and third regression models represent the first, second, and third regression models, respectively. Based on the first regression model and the deviated well drilling sequence data, the lithology LITH can be calculated; based on the second regression model and the deviated well drilling sequence data, the rock mass compressive strength RC can be calculated; and based on the second regression model and the deviated well drilling sequence data, the rock mass integrity KV can be calculated.

[0030] Historical data is constructed using collected rock samples and drilling sequence data at the sample locations. The drilling sequence data at the sample locations is used as input, and the lithology, rock mass compressive strength, and rock mass integrity corresponding to the rock samples are used as labels to train a first regression model, a second regression model, and a third regression model.

[0031] The evaluation model can use a weighted summation method, applying different weights to lithology (LITH), rock mass compressive strength (RC), and rock mass integrity (KV). The evaluation score is calculated, and the relationship between the evaluation score and the surrounding rock grade is determined. When the evaluation score falls within a certain score range, the surrounding rock grade corresponding to that score range is taken as the surrounding rock grade RANK at that location.

[0032] In another embodiment, in order to obtain rock mechanics parameters characterizing the rock mechanics properties of the deviated well based on the deviated well drilling sequence data, a first identification model can be pre-trained, and the deviated well drilling sequence data can be converted into rock mechanics parameters based on the first identification model.

[0033] See Figure 2 The first pre-trained recognition model of this invention employs a multi-target recognition model, comprising an input layer, an encoding layer, a first multi-tower network layer, a memory layer, a second multi-tower network layer, a decoding layer, and an output layer. The input layer uses the input deviated well drilling sequence data and transmits the deviated well drilling sequence data to the coding layer; The encoding layer is used to encode the deviated well drilling sequence data into a first feature vector, and then transmit the first feature vector to multiple tower networks in the first multi-tower network layer: Where T, POS, DP, TOR, RSP, DSP, and DIS represent time, position, drilling pressure, torque, rotational speed, drilling speed, and displacement, respectively, and the coding layer adopts the seq2seq sequence model.

[0034] The first multi-tower network layer comprises three tower networks, used to process the first feature vector to generate three intermediate vectors corresponding to lithology (LITH), rock mass compressive strength (RC), and rock mass integrity (KV) in the initial model data, and to transmit the three intermediate vectors to the memory layer; each tower network consists of N layers of LSTM neural networks:

[0035]

[0036]

[0037] Where N is the number of layers, which can be 2-4.

[0038] The memory layer is used to remember the three intermediate vectors corresponding to time tT to t, respectively, to obtain the memory vector, and then input the memory vector into the second multi-tower network layer: Where T is the memory cycle, and the memory layer can be a key-value network.

[0039] The second multi-tower network layer also includes three tower networks, used to process the memory vectors and generate three decoded vectors corresponding to the lithology LITH, rock mass compressive strength RC, and rock mass integrity KV in the initial model data, and transmit the three decoded vectors to the decoding layer; each tower network consists of N layers of LSTM neural networks:

[0040]

[0041]

[0042] Where N is the number of layers, which can be 2-4.

[0043] The decoding layer comprises four decoding sub-modules. The first decoding sub-module processes the first decoding vector to obtain the lithology LITH. The second decoding sub-module processes the second decoding vector to obtain the rock mass compressive strength RC. The third decoding sub-module processes the third decoding vector to obtain the rock mass integrity KV. The fourth decoding sub-module processes the first, second, and third decoding vectors to obtain the surrounding rock grade RANK.

[0044]

[0045]

[0046]

[0047] in, , , These correspond to lithology (LITH), rock mass compressive strength (RC), rock mass integrity (KV), and surrounding rock grade (RANK), respectively; First decoding submodule Fourth decoding submodule Using a classification model, the first decoding submodule Fourth decoding submodule A generative model is used.

[0048] The fourth decoding submodule is used to process the first, second, and third decoding vectors to obtain the surrounding rock grade RANK, specifically including: The first decoded vector, the second decoded vector, and the third decoded vector are concatenated to obtain the first concatenated vector: ; The fourth decoding submodule is used to decode the first concatenated vector to obtain the surrounding rock grade: .

[0049] The output layer is used to output the results obtained from the decoding layer.

[0050] By using the first identification model based on a multi-object identification model, the rock mechanics parameters of the deviated well can be obtained quickly and accurately based on the deviated well drilling sequence data, thus improving efficiency and accuracy.

[0051] Similarly, after constructing the first identification model, historical data is needed to train it. This historical data is also constructed using collected rock samples and drilling sequence data at the sample locations. The drilling sequence data at the sample locations is used as input, with the lithology, rock mass compressive strength, rock mass integrity, and surrounding rock grade corresponding to the rock sample as labels, respectively. The first identification model is then trained, where lithology, rock mass compressive strength, and rock mass integrity can be based on experimental measurements, while the surrounding rock grade is given by expert experience or through an evaluation model.

[0052] S3: Generate the inclined shaft BIM model based on the initial model data.

[0053] Using initial model data, a BIM model corresponding to the inclined shaft is generated. The inclined shaft BIM model can be generated solely from initial model data reflecting the rock mechanics properties of the inclined shaft, or it can be generated from initial model data reflecting the rock mechanics properties, first model data reflecting the geometric properties of the inclined shaft, and second model data reflecting the geological properties of the inclined shaft. The first and second model data can be transmitted to the integrated control platform all at once after the advanced exploration body conducts advanced exploration of the inclined shaft, or they can be transmitted to the integrated control platform as streaming data during the advanced exploration body's exploration of the inclined shaft.

[0054] Generate a BIM model of the inclined shaft based on the initial model data, specifically including: S31: Obtain first model data and second model data, wherein the first model data is constructed based on the geometric parameters of the inclined shaft and the second model data is constructed based on the geological parameters of the inclined shaft; S32: Use the initial model data as the first layer data of the BIM model, use the first model data as the second layer data of the BIM model, and use the second model data as the third layer data of the BIM model. S33: Overlay the data from the first layer, the second layer, and the third layer to generate the inclined shaft BIM model.

[0055] S4: Send the inclined shaft BIM model to the client for display. Displaying the inclined shaft BIM model on the client specifically includes: obtaining the surrounding rock grade of the inclined shaft; and displaying the inclined shaft BIM model using different types of display methods according to the surrounding rock grade; these different types of display methods include, but are not limited to, different colors. Areas with high surrounding rock grades are displayed using light colors (such as green), while areas with low surrounding rock grades are displayed using dark colors (such as red). Users can zoom in, zoom out, rotate, and display the inclined shaft BIM model in different ways by dragging or sliding the mouse.

[0056] In one embodiment, the first model data and the second model data are generated by the advanced probe after it performs advanced detection of the inclined shaft. A second recognition model and a third recognition model are deployed on the integrated control platform. After obtaining the 3D point cloud data (3DPoint) and imaging data (Image) collected by the advanced probe, the integrated control platform uses the second and third recognition models to identify the 3D point cloud data (3DPoint) and imaging data (Image) respectively, thereby obtaining the first model data and the second model data. The advanced probe performs advanced detection of the inclined shaft, achieving rapid 360° scanning detection of the inclined shaft using a 3D laser scanning device mounted on the advanced probe, thereby obtaining the 3D point cloud data (3DPoint), and obtaining the imaging data (Image) within the inclined shaft using a high-definition camera device mounted on the advanced probe.

[0057] The 3D point cloud data 3DPoint is input into the second recognition model to obtain the first model data, specifically: Acquire 3D point cloud data collected by the advanced probe; A pre-trained second recognition model is used to identify 3D point cloud data and obtain the geometric parameters of the inclined shaft. These parameters include at least the location and contour of the inclined shaft. The pre-trained second recognition model employs the ConvPoint model, a convolutional neural network model based on consecutive convolutional layers; alternatively, models such as PointNet, PointCNN, and KPConv can also be used.

[0058] in The geometric parameters of the inclined shaft include at least the location POS and the inclined shaft profile corresponding to that location POS; The first model data is constructed based on the geometric parameters of the inclined shaft. The first model data includes the location POS and the inclined shaft profile CONT.

[0059] The image data (Image) is input into the third recognition model to obtain the second model data, specifically: Acquire imaging data collected by the advanced probe; The pre-trained third recognition model identifies imaging data and obtains the geological parameters of the inclined well, including the geological body type, occurrence, and lithology at different locations. The pre-trained third recognition model uses the stacked convolutional neural network model CNN_STACKED, but various modified convolutional neural networks can also be used.

[0060] in Geological parameters for the deviated well include at least its location and the type, occurrence, and lithology of the geological body at that location; The second model data is constructed based on the geological parameters of the inclined shaft; the second model data includes location POS, geological body type GB_CAT, geological body occurrence GB_ATT, and lithology LITH.

[0061] In another embodiment, the first and second model data are obtained as streaming data by the advanced detection body during the advanced detection of the inclined shaft. A second and third recognition model are deployed during the advanced detection. After acquiring 3D point cloud data (3Dpoint) and imaging data (Image), the advanced detection body processes the 3D point cloud data (3Dpoint) and imaging data (Image) using its internally deployed second and third recognition models to obtain the first and second model streaming data, which are then transmitted to the integrated control platform in real time. The integrated control platform first transmits the initial model data as the first layer data of the inclined shaft BIM model to the client for display. Upon receiving the first and second model streaming data transmitted by the advanced detection body, the integrated control platform transmits the first model streaming data as the second layer data of the inclined shaft BIM model and the second model streaming data as the third layer data of the inclined shaft BIM model to the client for display. This allows for simultaneous advanced detection and visual management, achieving real-time visualization of the inclined shaft advanced detection process, intuitively presenting the geological conditions of the inclined shaft, facilitating rapid assessment of construction risks by relevant personnel, and enabling timely risk prevention and control measures.

[0062] To improve the efficiency and accuracy of advanced detection, this invention also proposes an adaptive control method for the advanced detection volume using the initial model data, first model stream data, and second model stream data obtained above. This method is applied to the advanced detection volume and specifically includes: SA: Obtain initial model data of the inclined shaft from the integrated control platform. After the integrated control platform processes and obtains the initial model data, and before the advanced probe performs advanced exploration of the inclined shaft, the advanced probe sends an acquisition request to the integrated control platform. The acquisition request includes the inclined shaft identifier. Based on the inclined shaft identifier, the initial model data corresponding to the inclined shaft identifier is obtained.

[0063] SB: Acquire the detection data collected by the detection equipment, process the detection data to obtain the first model stream data and the second model stream data.

[0064] SC: Merge the first model stream data, the second model stream data, and the initial model data to obtain the first BIM model data. Based on the location POS, correlate and merge the first model stream data, the second model stream data, and the initial model data to obtain the first BIM model data X. BIM Location (POS), surrounding rock grade (RANK), lithology (LITH), integrity (KV), deviated shaft profile (CONT), geological body category (GB_CAT), geological body occurrence (GB_ATT).

[0065] SD: Input the first BIM model data into the driving model to obtain the first driving data. Specifically, this involves: acquiring a pre-trained driving model; acquiring second driving data at multiple moments of the advanced detection body; and inputting the first BIM model data and the second driving data at multiple moments into the driving model to generate the first driving data.

[0066] In one embodiment, the advanced detection body employs a tracked displacement device, therefore the drive data uses velocity SP, acceleration ACC, and azimuth ORI, i.e., the drive data can be expressed as:

[0067] In another embodiment, the advanced detection body employs a four-wheel or multi-wheel displacement device, therefore the first driving data also includes the azimuth angle of each wheel, i.e., the driving data can be expressed as: ,in Let be the azimuth angle of the k-th wheel at time t.

[0068] The driving model adopts an encoder-decoder architecture. The encoder includes a first encoding module and a second encoding module. The first encoding module encodes the first BIM model data to obtain a first encoding vector. Since different data in the first BIM model data have varying impacts on the detection, different weights are used for different feature data. For example, higher weights are applied to surrounding rock grade (RANK), lithology (LITH), integrity (KV), shaft outline (CONT), and geological body category (GB_CAT), while lower weights are applied to location (POS) and geological body attitude (GB_ATT). The first encoding module uses a convolutional neural network model based on an attention mechanism. .

[0069] The second encoding module is used to encode the second driving data at multiple time points to obtain the second encoding vector. The second encoding module adopts the BILSTM model.

[0070] , where n can be obtained based on the actual calculation accuracy and efficiency.

[0071] Decoder is used for and The concatenated vector is decoded to obtain the first driving data, specifically... Decoder BI_LSTM network.

[0072] Similarly, when the advanced detection body uses a tracked displacement device, the predicted first driving data can be expressed as:

[0073] When the advanced detection body uses a four-wheel or multi-wheel displacement device, the predicted first driving data can be expressed as: ,in To predict the azimuth angle of the k-th wheel at time t+1.

[0074] SE: Drives the advanced probe to move based on the first driving data. The driving data generated by the driving model can drive the advanced probe to move slowly at locations with low surrounding rock grade, poor integrity, incomplete well outline (e.g., protrusions and depressions), geological intrusions, and fractures. It can also drive the advanced probe to move rapidly at locations with high surrounding rock grade, good integrity, intact well outline, and little or no lithological change.

[0075] SF: Update the initial model data using the first BIM model data; specifically, use the first BIM model data and the initial model data at the current moment together as the initial model data for the next moment.

[0076] SG: Repeat steps SB-SF until the advance exploration of the deviated shaft is completed.

[0077] See Figure 3 The present invention also proposes a deviated well data processing system, which includes: The acquisition module is used to acquire deviated well drilling sequence data; The processing module is used to process the deviated well drilling sequence data to obtain initial model data; the initial model data includes at least location, lithology, rock mass compressive strength, rock mass integrity, and surrounding rock grade; The generation module is used to generate a BIM model of the inclined shaft based on the initial model data; The sending module is used to send the inclined shaft BIM model to the client for display.

[0078] The processing module processes the deviated well drilling sequence data to obtain initial model data. Specifically, it includes: inputting the deviated well drilling sequence data into a pre-trained first recognition model, and outputting the initial model data based on the pre-trained first recognition model. The pre-trained first recognition model adopts a multi-target recognition model.

[0079] The multi-target recognition model includes an input layer, an encoding layer, a first multi-tower network layer, a memory layer, a second multi-tower network layer, a decoding layer, and an output layer. The input layer is used to input the deviated well drilling sequence data and transmit the deviated well drilling sequence data to the encoding layer; The encoding layer is used to encode the deviated well drilling sequence data into a first feature vector, and transmit the first feature vector to the first multi-tower network layer; The first multi-tower network layer includes three tower networks, which are used to process the first feature vector respectively to generate a first intermediate vector, a second intermediate vector and a third intermediate vector corresponding to lithology, rock mass compressive strength and rock mass integrity, and transmit the first intermediate vector, the second intermediate vector and the third intermediate vector to the memory layer; The memory layer is used to remember the first intermediate vector, the second intermediate vector and the third intermediate vector corresponding to multiple time points, to obtain a memory vector, and then transmits the memory vector to the second multi-tower network layer; The second multi-tower network layer includes three tower networks, which are used to process the memory vectors respectively to generate a first decoding vector, a second decoding vector, and a third decoding vector corresponding to lithology, rock mass compressive strength, and rock mass integrity, and transmit the first decoding vector, the second decoding vector, and the third decoding vector to the decoding layer; The decoding layer is used to decode the first decoding vector, the second decoding vector, and the third decoding vector to obtain the lithology, the rock mass compressive strength, the rock mass integrity, and the surrounding rock grade, and transmits the lithology, the rock mass compressive strength, the rock mass integrity, and the surrounding rock grade to the output layer; The output layer is used to output the lithology, the rock mass compressive strength, the rock mass integrity, and the surrounding rock grade.

[0080] The decoding layer includes four decoding sub-modules. Decoding the first decoding vector, the second decoding vector, and the third decoding vector to obtain the lithology, the rock mass compressive strength, the rock mass integrity, and the surrounding rock grade specifically includes: The first decoding vector is decoded using the first decoding submodule to obtain the lithology; The second decoding vector is decoded using the second decoding submodule to obtain the compressive strength of the rock mass; The third decoding vector is decoded using the third decoding submodule to obtain the integrity of the rock mass; The fourth decoding submodule is used to decode the first decoding vector, the second decoding vector, and the third decoding vector to obtain the surrounding rock grade.

[0081] In one embodiment, the acquisition module is also used to acquire three-dimensional point cloud data and imaging data collected by the advanced probe.

[0082] The processing module is also used to identify the three-dimensional point cloud data and imaging data based on the pre-trained second recognition model and the pre-trained third recognition model, respectively, to obtain the inclined well geometric parameters and inclined well geological parameters, and to construct the first model data and the second model data based on the inclined well geometric parameters and the inclined well geological parameters, respectively. The inclined well geometric parameters include at least the location and the inclined well outline, and the inclined well geological parameters include the geological body type, occurrence and lithology at different locations.

[0083] The generation module generates an inclined shaft BIM model based on the initial model data, specifically including: acquiring first model data and second model data; using the initial model data as the first layer data of the BIM model, using the first model data as the second layer data of the BIM model, using the second model data as the third layer data of the BIM model, and overlaying the first layer data, the second layer data and the third layer data to generate the inclined shaft BIM model.

[0084] In another embodiment, the acquisition module is further configured to acquire the first model stream data and the second model stream data sent by the adaptive advanced probe, the generation module is further configured to generate the first layer data of the BIM model based on the initial model data, generate the second layer data of the BIM model based on the first model stream data, and generate the third layer data of the BIM model based on the second model stream data, and the sending module is further configured to send the first layer data to the client for display, and then send the second layer data and the third layer data to the client for display in real time.

[0085] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., high-density digital video discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).

[0086] The memory in this embodiment of the invention can be volatile memory or non-volatile memory, or may include both. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which serves as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM). It should be noted that the memory used in the methods described in this invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0087] In implementation, each step of the above method can be completed by integrated logic circuits in the processor's hardware or by instructions in software. The steps of the method disclosed in the embodiments of this application can be directly implemented by a hardware processor, or by a combination of hardware and software modules in the processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, detailed descriptions are omitted here.

[0088] It should be noted that the processor in the embodiments of this application can be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method embodiments can be completed by the integrated logic circuitry in the processor's hardware or by instructions in software form. The processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied as being executed by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above methods.

[0089] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the foregoing application concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions claimed in this application.

Claims

1. A method for processing deviated well data, characterized in that, The method includes: Acquire deviated well drilling sequence data; the deviated well drilling sequence data includes at least time, location, drilling pressure, torque, rotational speed, drilling rate, and displacement; The drilling sequence data of the deviated well is processed to obtain initial model data; the initial model data includes at least location, lithology, rock mass compressive strength, rock mass integrity and surrounding rock grade. Generate a BIM model of the inclined shaft based on the initial model data; The BIM model of the inclined shaft is sent to the client for display.

2. The deviated well data processing method according to claim 1, characterized in that, The process of processing the deviated well drilling sequence data to obtain initial model data specifically includes: inputting the deviated well drilling sequence data into a pre-trained first recognition model, and outputting the initial model data based on the pre-trained first recognition model. The pre-trained first recognition model adopts a multi-target recognition model.

3. The deviated well data processing method according to claim 2, characterized in that, The multi-target recognition model includes an input layer, an encoding layer, a first multi-tower network layer, a memory layer, a second multi-tower network layer, a decoding layer, and an output layer. The input layer is used to input the deviated well drilling sequence data and transmit the deviated well drilling sequence data to the encoding layer; The encoding layer is used to encode the deviated well drilling sequence data into a first feature vector, and transmit the first feature vector to the first multi-tower network layer; The first multi-tower network layer includes three tower networks, which are used to process the first feature vector respectively to generate a first intermediate vector, a second intermediate vector and a third intermediate vector corresponding to lithology, rock mass compressive strength and rock mass integrity, and transmit the first intermediate vector, the second intermediate vector and the third intermediate vector to the memory layer; The memory layer is used to remember the first intermediate vector, the second intermediate vector and the third intermediate vector corresponding to multiple time points, to obtain a memory vector, and then transmits the memory vector to the second multi-tower network layer; The second multi-tower network layer includes three tower networks, which are used to process the memory vectors respectively to generate a first decoding vector, a second decoding vector, and a third decoding vector corresponding to lithology, rock mass compressive strength, and rock mass integrity, and transmit the first decoding vector, the second decoding vector, and the third decoding vector to the decoding layer; The decoding layer is used to decode the first decoding vector, the second decoding vector, and the third decoding vector to obtain the lithology, the rock mass compressive strength, the rock mass integrity, and the surrounding rock grade, and transmits the lithology, the rock mass compressive strength, the rock mass integrity, and the surrounding rock grade to the output layer; The output layer is used to output the lithology, the rock mass compressive strength, the rock mass integrity, and the surrounding rock grade.

4. The deviated well data processing method according to claim 3, characterized in that, The decoding layer includes four decoding sub-modules. Decoding the first decoding vector, the second decoding vector, and the third decoding vector to obtain the lithology, the rock mass compressive strength, the rock mass integrity, and the surrounding rock grade specifically includes: The first decoding vector is decoded using the first decoding submodule to obtain the lithology; The second decoding vector is decoded using the second decoding submodule to obtain the compressive strength of the rock mass; The third decoding vector is decoded using the third decoding submodule to obtain the integrity of the rock mass; The fourth decoding submodule is used to decode the first decoding vector, the second decoding vector, and the third decoding vector to obtain the surrounding rock grade.

5. The deviated well data processing method according to claim 1, characterized in that, Generate a BIM model of the inclined shaft based on the initial model data, specifically including: Acquire first model data and second model data, wherein the first model data is constructed based on the geometric parameters of the inclined shaft and the second model data is constructed based on the geological parameters of the inclined shaft; The initial model data is used as the first layer data of the BIM model, the first model data is used as the second layer data of the BIM model, and the second model data is used as the third layer data of the BIM model. The first layer data, the second layer data and the third layer data are superimposed to generate the inclined shaft BIM model.

6. The deviated well data processing method according to claim 5, characterized in that, The acquisition of the first model data specifically includes: Acquire 3D point cloud data collected by the advanced probe; The second recognition model, which is pre-trained, is used to identify three-dimensional point cloud data and obtain the geometric parameters of the inclined shaft. The geometric parameters of the inclined shaft include at least the location and the outline of the inclined shaft. The first model data is constructed based on the geometric parameters of the inclined shaft.

7. The deviated well data processing method according to claim 5, characterized in that, The acquisition of the second model data specifically includes: Acquire imaging data collected by the advanced probe; The pre-trained third recognition model is used to identify the imaging data and obtain the geological parameters of the inclined well. The geological parameters of the inclined well include the geological body type, occurrence and lithology at different locations. The second model data is constructed based on the geological parameters of the deviated well.

8. A deviated shaft data processing system, characterized in that, The system includes: The acquisition module is used to acquire deviated well drilling sequence data; The processing module is used to process the deviated well drilling sequence data to obtain initial model data; the initial model data includes at least location, lithology, rock mass compressive strength, rock mass integrity, and surrounding rock grade; The generation module is used to generate a BIM model of the inclined shaft based on the initial model data; The sending module is used to send the inclined shaft BIM model to the client for display.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium for storing computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the steps of the method according to any one of claims 1-7.

Images